Niagara Region Cancer Report Incidence, Mortality and Temporal Trends

Transcription

1 Niagara Region Cancer Report Incidence, Mortality and Temporal Trends PLANNING, RESEARCH, EVALUATION AND POLICY DEVELOPMENT UNIT Niagara Region Public Health Department October 2007

2 TABLE OF CONTENTS OBJECTIVE... 5 METHODS... 6 DATA SOURCES... 6 DATA LIMITATIONS... 7 STATISTICAL ANALYSIS... 7 SCREENING, DIAGNOSIS AND TREATMENT OF CANCER... 8 IDENTIFICATION OF EPIDEMIOLOGICAL MEASURES AND DEFINITIONS... 9 Incidence and Mortality... 9 Deaths to Cases Ratio... 9 Age... 9 Potential Years of Life Lost... 9 DEMOGRAPHICS OF THE NIAGARA REGION LIFESTYLE AND BEHAVIOUR RESULTS I. CURRENT INCIDENCE AND MORTALITY (2003) Table 1. New Cases and Deaths for Cancer Sites by Gender, Niagara Region, Figure 1.a Percentage distribution of new cases for selected cancer sites, males, Niagara Region, Figure 1.b Percentage distribution of deaths for selected cancer sites, males, Niagara Region, Figure 2.a. Percentage distribution of new cases for selected cancer sites, females, Niagara Region, Figure 2.b Percentage distribution of deaths for selected cancer sites, females, Niagara Region, II. TIME TRENDS IN INCIDENCE AND MORTALITY ( ) A. ALL CANCER SITES Males Figure 3 a. New cases and age-standardized incidence rates for all male cancers, Niagara Region ( ) Figure 3b. Deaths and age-standardized mortality rates for all male cancers, Niagara Region ( ) Females Figure 4a. New cases and age-standardized incidence rates for all female cancers, Niagara Region ( ) Figure 4b. Deaths and age-standardized mortality rates for all female cancers, Niagara Region ( ) B. Comparison of temporal trends for selected cancer sites in the Niagara Region Males Incidence Figure 5 a. Age-standardized incidence rates (ASIR) for selected cancer sites, males, Niagara Region, Mortality Figure 5 b. Age-standardized mortality rates (ASMR) for selected cancer sites, males, Niagara Region, Females Incidence Figure 6a. Age-standardized incidence rates (ASIR) for selected cancer sites, females, Niagara Region, Mortality Figure 6 b. Age-standardized mortality rates for selected cancer sites, females, Niagara Region,

3 C. AVERAGE ANNUAL PERCENTAGE CHANGE IN INCIDENCE AND MORTALITY OF SELECTED CANCERS (NIAGARA REGION, ) Table 2. Average annual percentage change in incidence and mortality (Niagara Region, ) D. COMPARISON OF NIAGARA REGION WITH PROVINCIAL TEMPORAL TRENDS IN INCIDENCE AND MORTALITY PROSTATE CANCER...26 Figure 7. Temporal trends of Prostate Cancer, , Niagara Region and Ontario BREAST CANCER Figure 8. Temporal trends for Breast Cancer, , Niagara Region and Ontario LUNG CANCER Figure 9. Temporal trends of Lung Cancer in females, , Niagara Region and Ontario. 31 Figure 10. Temporal trends in Lung Cancer in males, , Niagara Region and Ontario.. 31 COLORECTAL CANCER Figure 11. Temporal trends in Colorectal Cancer in females, , Niagara Region and Ontario Figure 12. Temporal trends in Colorectal Cancer in males, , Niagara Region and Ontario STOMACH CANCER Figure13. Temporal trends in Stomach Cancer in females , Niagara Region and Ontario Figure 14. Temporal trends in Stomach Cancer in males, , Niagara Region and Ontario PANCREATIC CANCER Figure 15. Temporal trends in Pancreatic Cancer in females, , Niagara Region and Ontario Figure 16. Temporal trends in Pancreatic Cancer in males, , Niagara Region and Ontario NON-HODGKIN S LYMPHOMA Figure 17. Temporal trends of Non-Hodgkin s Lymphoma in females, , Niagara Region and Ontario Figure 18. Temporal trends of Non-Hodgkin s Lymphoma in males, , Niagara Region and Ontario UTERINE CANCER Figure 19. Temporal trends in Uterine Cancer, , Niagara Region and Ontario OVARIAN CANCER Figure 20. Temporal trends in Ovarian Cancer, , Niagara Region and Ontario CANCER OF THE UTERINE CERVIX Figure 21. Temporal trends in Cancer of the Uterine Cervix, , Niagara Region and Ontario BLADDER CANCER Figure 22. Temporal trends in Bladder Cancer in females, , Niagara Region and Ontario Figure 23. Temporal trends in Bladder Cancer in males, , Niagara Region and Ontario MELANOMA Figure 24. Temporal trends in Melanoma in females, , Niagara Region and Ontario Figure 25. Temporal trends in Melanoma in males, , Niagara Region and Ontario E. AGE-STANDARDIZED INCIDENCE AND MORTALITY BY AGE GROUPS FOR ALL CANCERS (NIAGARA REGION AND ONTARIO, ) Figure 28 a. Temporal trends of age group 0-19 in the Niagara Region ( ) Figure 28 b. Temporal trends of age group 0-19 in Ontario ( ) Figure 29 a. Temporal trends of age group in the Niagara Region ( ) Figure 29 b. Temporal trends of age group in Ontario ( ) Figure 30 a. Temporal trends of age group in the Niagara Region ( ) Figure 30 b. Temporal trends of age group in Ontario ( ) Figure 31 a. Temporal trends of age group 70+ in the Niagara Region ( ) Figure 31 b. Temporal trends of age group 70+ in Ontario ( ) F. POTENTIAL YEARS OF LIFE LOST DUE TO CANCER Table 3. Potential Years of Life Lost

4 Figure 32. Potential Years of Life Lost in 2002 by Cause CONCLUSIONS REFERENCES DESCRIPTION OF ACRONYMS Appendix I International Classification of Disease Table 1: ICD 9 Cancer codes Table 2: ICD 10 Cancer codes Appendix II: Actual Data for New Cases and Deaths Table 1. Temporal Trends in Incidence for All Cancers Table 2. Temporal Trends in Mortality for All Cancers Appendix III: Temporal Data for New Cases and Deaths by Cancer Site Table 1. Prostate Cancer Table 2. Breast Cancer Table 3. Lung Cancer Table 4. Female Reproductive Cancers Table 5. Cancers of the Oral Cavity Table 6. Cancer of the Pancreas Table 7. Cancer of the Stomach Table 8: Colorectal Cancer Table 9. Kidney Cancer Table 10. Cancer of the Bladder Table 11. Endocrine Glands (Thyroid)* Table 12: Leukaemia Table 13: Hodgkin s Disease* Table 14. Non-Hodgkin s Disease...79 Table 15. Melanoma

5 OBJECTIVE The goal of this report is to present cancer incidence and mortality, temporal trends and age-specific information for males and females in the Niagara Region from 1986 to Such information provides a useful input to the development, implementation, and evaluation of health promotion, prevention, treatment and other policies for cancer control. This report was updated by Deborah Moore, Epidemiologist. Original report was created by Pia K. Muchaal, Epidemiologist, with assistance from Nicole Coffer and Angela McVittie, Planning, Research, Evaluation and Policy Unit. The presentation style for the data was adapted from Canadian Cancer Statistics

6 METHODS Data Sources The Ontario Cancer Registry is operated by Cancer Care Ontario (CCO), (which replaced Ontario Cancer Treatment and Research Foundation (OCTRF) in Cancer Care Ontario registers all newly diagnosed cases of cancer, except non-melanoma skin cancer. Although cancer is not a reportable disease in Ontario, the Cancer Act mandates CCO to record, compile and report cancer data (CCO, 2003). Cancer Care Ontario relies on information obtained primarily from four major data sources: 1. Hospital discharge summaries which include diagnosis of cancer; 2. Pathology reports with mention of cancer; 3. Records of patients referred to CCO s eight Regional Cancer Centres or the Princess Margaret Hospital; and 4. Death certificates with cancer as the underlying cause of death. All records, except pathology reports, are coded at the source and are submitted to CCO in machine-readable form. All hospital and private pathology laboratories send the pathology reports to CCO where staff code and enter the information into a computerized database. Duplicate records are identified and resolved via computerized record linkage and logic. The principle cancer data quality issues are consistency in classification and coding, completeness of registration, and validity of data recorded. Cancer sites are coded according to the International Classification of Diseases, Ninth Revision (ICD9) (WHO, 1976). Invasive neoplasms are included in the ICD9 codes ranging from 140 to 208. Most of these three-digit codes describe the cancer site but some describe the morphology instead. ICD10 codes were used from the Provincial Health Planning Database for the calculation of Potential Years of Life Lost in The ICD9 and ICD10 codes and their respective descriptions are outlined in Appendix I. Actual data for new cases and deaths presented in this report are included in Appendices II and III. 6

7 METHODS Data Limitations With few exceptions, variations in classification and coding practices generally have minimal effect on incidence rates. The main exceptions are in coding cancers of the lower sigmoid colon and recto-sigmoid junction to the colon (ICD 153) or rectum (ICD 154). The solution to this problem was to combine these sites into colorectal (ICD ). The International Classification of Diseases (ICD9) is not well-adapted for consistent coding on non-hodgkin s lymphomas. For this reason, the two codes for non- Hodgkin s lymphoma are combined (ICD9 200, 202). Similarly, oral cancers (ICD ), cancers of the brain and other central nervous system sites (ICD ), and the several types of leukemia (ICD ) are considered together. For analysis using ICD10 codes (PYLL), breast cancer is not specific to male or female as it is in ICD9. The classification of two other cancer sites was changed during the period considered in this report and will influence apparent trends in incidence rates. In 1988, the Cancer Registry began to code borderline epithelial neoplasms of the ovary as malignancies. It is estimated that 10-15% of ovarian cancers are in this category. In 1989 classification was also changed for invasive papillary transitional carcinomas of the bladder. While these tumours are potentially malignant, they have not yet become invasive. The effect of this change was an apparent drop in incidence of approximately 25%. As a result of the change in coding it is inappropriate to compare bladder cancer rates before and after Completeness of case ascertainment is the proportion of all incident cancers in the population included in the registry. Some cancers are not diagnosed, but this occurs infrequently in Ontario. Incompleteness is more likely as a result of gaps in case-finding procedures. Cancer Care Ontario uses percent death certificate only (%DCO) and the mortality/incidence ratio (%M/I), in addition to statistical techniques to estimate completeness of ascertainment. A death certificate only (DCO) case is one for which a death certificate is the only source of information about the diagnosis of cancer. The overall OCR %DCO rates are less than 2%. For cancers that are managed on an outpatient basis (lip and melanoma), %DCO is approximately 11%. Statistical Analysis SEERStat, a statistical software package developed for use with the Surveillance, Epidemiology, and End Results (SEER) Registry data 1, and modified to fit the Ontario Cancer Registry incidence and mortality data, was used to calculate cancer incidence, mortality, average annual percent change, and trends in the Niagara Region and Ontario. For the average annual percent change statistics, weighted least squares were used to complete the calculation. Cancer Care Ontario provided the Regional Niagara Public Health Department with a copy of the SEERStat software. The Potential Years of Life METHODS 1 A programme of the United States National Cancer Institute 7

8 Lost (PYLLs) were determined based on data from Provincial Health Planning Database where ICD10 codes were used to classify disease. Screening, Diagnosis and Treatment of Cancer Traditionally the aim of prescriptive screening is to detect disease in its initial stages in presumptively healthy members of a defined population (Last, 2001). The characteristics required for a screening test are: adequate sensitivity and specificity; evidence, ideally from randomized clinical trials, of lowered mortality of those undergoing the screening, and acceptance of the test by the target population (Armstrong, 2002). Furthermore, the cost of an on-going screening programme must be economically feasible and adequate follow-up diagnostic and treatment services must be available for the particular cancer (Armstrong, 2002). Therefore, not all cancers are easily screened, and not all screening tests are suitable for use in population-wide screening programmes. Prior to 1997, the Ontario Cancer Treatment and Research Foundation (OCTRF) and the Ontario Cancer Institute/Princess Margaret Hospital (OCI/PMH) represented the formal cancer care system in Ontario. The OCTRF s primary responsibility was to deliver specialized treatment services (radiation and systemic therapy). Since assuming responsibility of the OCTRF in 1997, Cancer Care Ontario, as part of their mandate, have implemented a provincial cancer framework which is delivered through regional networks. CCO collaborates with all cancer stake-holders to create an integrated cancer control system. Cancer treatment services are provided by the regional cancer centres (RCC), host hospitals, and community hospitals. The regional cancer centres deliver outpatient services and are responsible for 75% of the radiation treatment in the province, 50% of chemotherapy, and programmes in supportive care, prevention, screening, education and research. Host hospitals provide inpatient services including surgery and supportive care, diagnostic imaging, pathology and laboratory services and undertake research. Community hospitals take on a similar role as host hospitals (CCO, 2001). Until the Niagara Region cancer centre is completed, cancer patients resident in the Region must travel to Hamilton and New York for radiation treatment. 8

9 METHODS Identification of Epidemiological Measures and Definitions Incidence and Mortality Cancer incidence is defined as the number of new cases identified in the Niagara Region during a specified calendar period and reported to the OCR. Cases with unknown age are included in the totals, but excluded from the individual age-specific groups. Due to the few pediatric cases (0-14 years) of cancer, specific statistics are not presented for the three youngest groups, but they are included in the Niagara totals. Cancer data for children and youth 0-19 was pooled and presented as a summary. Cancer deaths are identified based on the underlying cause of death as reported by the certifying physician on the medical certificate of death. Information on deaths is obtained from the Office of the Registrar General of Ontario. Incidence and mortality were calculated per 100,000 people per annum. Rates for the Niagara Region and Ontario were standardized based on the 1991 Canadian population. Deaths to Cases Ratio The mortality/incidence ratio (%M/I) is the number of deaths due to cancer in a given time period divided by the number of incident cases for the same period, expressed as a ratio or percent. The ratio is inversely associated with survival and can only be interpreted in the context of survival rates. Age The age of the patient at the time of diagnosis in completed years. The age-specific rates in this report were calculated for ages 0-19, 20-49, and 70+. Potential Years of Life Lost The Potential Year of Life Lost (PYLL s) for 2002 is the ratio for a given period of the total years of life lost before age 75 to the total population under 75 years of age by ICD chapter (per 1,000). 9

10 DEMOGRAPHICS OF THE NIAGARA REGION According to the 2001 Statistics Canada census, there were 410,574 residents living in the Niagara Region, a 1.8% increase in the overall population compared to 5 years prior. The median age of the population was 40 years with 17% of the population over the age of 65; 42.8% and 57.2% are men and women respectively. Over 52% percent of the Niagara Region respondents in the 2001 census identified themselves as Canadian or British. Italians comprised 11.6% of the sample, French 4.1%, and German 5.7%. The remaining 26% identified as being of an ethnic origin other than those mentioned above. While the general demographic distribution of the Niagara Region parallels the provincial profile, the population of Ontario increased by 6% between 1996 and 2001, and the proportion of people over the age of 65 comprised 13% of the population. The inherent impact of the slower rate of population growth and the increased proportion of people over the age of 65 is a general increase in cancer incidence and mortality. Lifestyle and Behaviour Tobacco use, diet (including alcohol consumption) and lack of physical activity are associated with over 50% of all cancer deaths in the industrial world (Targeting Cancer, 2003). In general, smoking tobacco accounts for 90% of lung cancers and 30% of all cancers could be prevented by not smoking (WCRF-AICR, 1997;Peto et al, 2004; AICR, 2007). In Ontario, with the implementation of the Smoke Free Ontario Act in 2006, the elimination of smoking in public places may lead to a potential decrease in number of smokers, most likely reducing the incidence and mortality from lung cancer. Alcohol consumption is a well-established cause of cancer of the upper aero-digestive tract (mouth, pharynx, larynx and oesophagus), liver and breast (WCRF, 2006). Dietary factors also contribute significantly to the effect on the risk of cancer. Approximately one-third of all cancers are directly related to diet (WCRF-AICR, 1997; AICR, 2007; Rohan et. al., 2007) and a cohort study completed by Calle et al states that cancer death rates were 52% higher in men and 62% higher in women among the heaviest participants compared with those of normal weight. The dietary factors for which there is strong evidence of linkages to cancer are: insufficient intake of vegetables, fruit and wholegrain cereals; and high intake of alcohol. There is convincing data that a diet high in fruits and vegetables protects against cancers of the mouth, pharynx, oesophagus, stomach and lungs, and there is indication that they are also protective against cancers of the larynx, pancreas, breast and bladder (WCRF-AICR, 1997; AICR, 2007). Vegetables also reduce the risk of colorectal cancer (WCRF-AICR, 1997). A lack of exercise and being overweight also increase the risk of cancer. In a comprehensive prospective investigation ( ) involving almost one million participants in the United States, 4.2%-19.8% of deaths from cancer were attributable to obesity (Calle et al., 2003). Conversely, physical activity has been identified as a riskmitigator of colon cancer and may also reduce the risk of breast and lung cancers (WCRF-AICR, 1997). DEMOGRAPHICS OF THE NIAGARA REGION 10

11 A behavioural and lifestyle profile of the Niagara Region residents was determined through the Rapid Risk Surveillance Report (2006), a self-reporting survey conducted on 1202 Niagara residents 18 years of age and older. In this report, 39% of the respondents (N=1098) had a body mass index within the normal range ( ) 34% of people were identified as overweight (BMI ) and 16.5% were obese (BMI 30+). Two trends were noted, body mass index increased with age and men were more likely to be overweight than women. At the time of the survey (2005) 21% of those surveyed smoked, with 17% of them reporting daily use of tobacco. Thirty-one percent of all respondents had never smoked. Although there were no differences in the proportion of male and female smokers, there were variations according to education and income levels. Of those with a postsecondary degree, 39.8% (N=625) were currently smoking. This proportion increased to 45% for those with only a high school degree, but decreased to 35.6% for individuals who had not completed high school. As income increases percentage of everyday smoking decreases (N=625). Of those who reported an income less than $30,000, 40% smoke everyday. For those who make between $30,000 and $70,000, 34% smoke everyday, which drops to 28% for those who make more than $70,000. Of the 485 (N=841) individuals who reported consuming alcohol at least one day per week 67% reported consuming 1-2 drinks per day of drinking, while 7% consumed 5 or more drinks per drinking day. 11

12 I. Current Incidence and Mortality (2003) A total of 2274 cases of cancer were diagnosed in the Niagara Region in the year For both men and women 3 cancers comprised over half of the registrations. Of the total cancers in males, 57% of the cancers were prostate (28%), lung (15%), and colorectal (14%) (Table 1 and Figure 1.a). Similarly, 53% of all cancer in women were attributable to breast (28%), lung (13%), or colorectal cancers (12%) (Table 1 and Figure 2.a). These 3 cancers also account for over 50% of all cancer deaths in both men (N=566) and women (N=488) in For both males and females, lung cancer comprised about one quarter of all cancer deaths (Figure 2.b & 1.b). Table 1. New Cases and Deaths for Cancer Sites by Gender, Niagara Region, New Cases 2003 Deaths 2003 Cancer Total Male Female Total Male Female Lung Prostate Breast 292 * Colorectal Bladder Uterus Cervix * - * Non-Hodgkins Lymphoma Kidney Ovary Pancreas Stomach Leukaemia Oral Cavity Melanoma Brain Oesophagus Larynx * Testis * * - Thyroid * * 0 All other sites Total * Insufficient data Data Source: SEERStat, Not applicable 12

13 Figure 1.a Percentage distribution of new cases for selected cancer sites, males, Niagara Region, 2003 (N=1169). All other sites, 8.1 Thyroid, 0.5 Testis, 1.1 Larynx, 1.0 Oesophagus, 1.6 Brain, 1.4 Melanoma, 3.2 Lung, 14.6 Oral Cavity, 2.4 Leukaemia, 3.7 Stomach, 2.3 Pancreas, 1.9 Kidney, 5.1 Prostate, 27.7 Non-Hodgkins Lymphoma, 5.5 Bladder, 5.7 Colorectal, 14.1 Data Source: SEERStat, 2006 Figure 1.b Percentage distribution of deaths for selected cancer sites, males, Niagara Region, 2003 (N=566) All other Sites, 15.0 Larynx, 1.1 Lung, 28.6 Oesophagus, 3.5 Brain, 1.4 Melanoma, 1.1 Oral Cavity, 1.8 Leukaemia, 4.1 Stomach, 3.9 Pancreas, 3.7 Prostate, 12.2 Kidney, 4.4 Non-Hodgkins Lymphoma, 3.0 Bladder, 3.7 Colorectal, 12.5 Data Source: SEERStat,

14 Figure 2.a. Percentage distribution of new cases for selected cancer sites, females, Niagara Region, 2003 (N=1045) All other sites, 4.6 Thyroid, 2.8 Oesophagus, 0.8 Brain, 1.7 Melanoma, 2.7 Oral Cavity, 1.6 Lung, 13.3 Leukaemia, 4.3 Stomach, 1.1 Pancreas, 2.2 Ovary, 5.5 Kidney, 2.8 Breast, 27.7 Non-Hodgkins Lymphoma, 6.9 Cervix, 2.3 Uterus, 5.5 Bladder, 2.4 Colorectal, 12.0 Data Source: SEERStat, 2006 Figure 2.b Percentage distribution of deaths for selected cancer sites, females, Niagara Region, 2003 (N=488). All other Sites, 11.9 Oesophagus, 1.0 Brain, 3.5 Melanoma, 1.0 Oral Cavity, 1.2 Lung, 23.8 Leukaemia, 5.7 Stomach, 1.8 Pancreas, 3.7 Ovary, 4.5 Kidney, 1.8 Breast, 19.9 Non-Hodgkins Lymphoma, 6.4 Uterus, 2.3 Bladder, 2.0 Colorectal, 9.4 Data Source: SEERStat,

15 II. Time Trends in Incidence and Mortality ( ) A. All Cancer sites Males The number of cases and deaths of a disease provides an indication of the burden of illness on the health care system and the demands on the communities. Between 1986 and 2003 the overall number of cancer registrations of males increased nearly 50% from 831 to 1208 cases with an estimated annual change in rates of 0.1% (CI = -0.3, 0.4) (Figure 3a ). In early 1990 s the age-standardized rates peaked at 497 cases per 100,000, while the lowest rates (453 cases per 100,000) occurred in (Figure 3a). The increase in incidence during this period is at least partly due to the detection of prostate cancers following trans-urethral resection of the prostate (TURP) for suspected benign prostatic hypertrophy. The second rise in cases in the late 1990s most likely reflects the aging population of the Niagara Region. Cancer deaths over the same period range from 438 to 641, the mortality rate as identified by the annual percentage change significantly decreased (AAPC = -0.7%; CI = -1.3, -0.1) (Figure 3b). The agestandardized rates peaked in 1988 (270.8 deaths per 100,000) and decreased to 216 deaths per 100,000 in The 35% increase in cancer deaths in conjunction with a decreasing age-standardized mortality rate mirrors the demographic shifts in the population. 15

16 Figure 3 a. New cases and age-standardized incidence rates for all male cancers, Niagara Region ( ) 1,400 1,200 Age-standardized rates per 100,000 1, Source: SEERStat, 2006 ASIR New Cases Rates Standardized to 1991 Canadian Population Figure 3b. Deaths and age-standardized mortality rates for all male cancers, Niagara Region ( ) Age-standardized mortality rates per 100, Data Source: SEERStat, 2006 ASMR Deaths Rates Standardized to 1991 Canadian Population 16

17 Females The number of cancer cases in females increased by 47% between 1986 and 2003 and the age-standardized rates were slightly raised ( cases per 100,000; AAPC = 0.5; CI = 0, 0.9) (Figure 4a). The increase in cases combined with the slight increase of rates during this period may be a reflection of an aging population, and highlights the demands on the health care system. The number of deaths in females increased 47% between 1986 and 2003 and the mortality rate significantly increased ( deaths per 100,000) with an AAPC of 1.2; CI = 0.9, 1.5, emphasizing the demographic trends of an aging population. Figure 4a. New cases and age-standardized incidence rates for all female cancers, Niagara Region ( ) Age-standardized rates per 100, Data Source: SEERStat, 2006 ASIR New Cases Rates Standardized to 1991 Canadian Population 17

18 Figure 4b. Deaths and age-standardized mortality rates for all female cancers, Niagara Region ( ) Age-standardized mortality rates per 100, Data Source: SEERStat, 2006 ASMR Deaths Rates Standardized to 1991 Canadian Population 18

19 B. Comparison of temporal trends for selected cancer sites in the Niagara Region Trends of incidence and mortality rates between 1986 and 2003 for selected sites are shown for men in Figure 5.a and Figure 5.b and for women in Figures 6.a and 6.b. Average annual percent changes (AAPC) for the cancer sites are presented in section C. Males Incidence Among men, only prostate cancer and Non-Hodgkin s lymphoma showed significant increasing trends between 1986 and The standardized incidence of cancer of the prostate increased steadily in the Niagara Region at an average annual rate of approximately 1.8%, from 80 cases per 100,000 (1986) to 119 cases per 100,000 (2003). The increase in incidence prior to 1990 is at least partly due to the detection of cancers following trans-urethral resection of the prostate (TURP) for suspected benign prostatic hypertrophy. Between 1988 and 1993, the rate of prostate cancer increased by 60%. This spike in cases diagnosed is most likely attributable to detection as a result of the Canada-wide introduction of prostate-specific antigen test in Since 1997 rates have not varied significantly. Between 1986 to 2003, the incidence of Non-Hodgkin s lymphoma, the only other cancer in men besides prostate cancer showing a significant escalation in incidence, increased from 14 cases per 100,000 to 26 cases per 100,000 (AAPC = 1.8; CI = 0.4, 3.1). Rates for other cancers sites in men in the Niagara Region have remained constant or have otherwise declined. In contrast to trends in women, in men, lung cancer incidence rates have been decreasing. From 1986 to 2003 the decrease has been approximately 31% (AAPC = -1.6; CI = -2.4, -0.90). Colorectal cancer has remained relatively stable with no significant change in incidence. When assessing colon and rectal cancer independently, there is still no significant change, although the trend seems to be decreasing for both. The incidence of stomach cancer has remained constant between 1986 and 2000 averaging 15 cases yearly per 100,000, but dropped to approximately 10 cases per 100,000 between 2001 and Overall there has been a significant decrease in stomach cancer incidence (AAPC = -2.7; CI = -4.1, -1.2). 19

20 Figure 5 a. Age-standardized incidence rates (ASIR) for selected cancer sites, males, Niagara Region, Age-standardized incidence rates/100, Year Data Source: SEERStat, 2006 Stomach Lung Prostate Colorectal Non-Hodgkins Lymphoma Rates Standardized to 1991 Canadian Population 20

21 Mortality Despite the prominent increase in incidence of prostate cancer and Non-Hodgkin s lymphoma (NHL) there only appears to be a marginally significant increase in the NHL mortality, where yearly rates doubled between 1986 and 2003 (AAPC = 2.5; CI = 0.1, 5.1). Deaths from prostate cancer have remained stable during the same period, at approximately 30 cases per 100,000 per year. Mortality rates of lung cancer, colorectal cancer, and stomach cancer have been declining. A reduction of 1.9% (CI = -3.1, -0.07) per year has occurred in lung cancer rates. Overall, deaths from colorectal cancer have fallen, on average, by 1.6% annually (CI = -2.6, -0.6) and stomach cancer rates continue to show a prominent decline of approximately 2.8% (CI = -4.9, -0.7) per year. Figure 5 b. Age-standardized mortality rates (ASMR) for selected cancer sites, males, Niagara Region, Age-standardized mortality rates/100, Year Data Source: SEERStat, 2006 Stomach Lung Prostate Colorectal Non-Hodgkin's Lymphoma Rates Standardized to 1991 Canadian Population 21

22 Females Incidence Among women in the Niagara Region, only Lung cancer and Non-Hodgkin s Lymphoma had a significant increase in incidence between 1986 and Lung cancer incidence rates increased at an average of 1.3% per annum (CI = 0, 2.6) with an increase of 34 cases per 100,000 to a peak of 58 cases per 100,000 over this time period. Previous data showed an increase of 2.5% between 1979 and 2000, therefore overall, the lung cancer rates may be decreasing as we go further past the year 2000, but comparison must be taken with caution due to the differences in data between the two time periods. An increase occurred in the incidence rates of Non-Hodgkin s Lymphoma (AAPC = 2.2; CI = 0.1, 4.3) with an increase of 9 cases per 100,000 to a peak of 23 cases per 100,000 over this time period. Although the rates from 1979 to 2000 are not comparable with rates from 1986 to 2003, previous findings of thyroid cancer are noteable. Thyroid cancer comprised only 0.1% of all new cancer cases in the last 21 years in the Niagara Region, but there was a 156% increase in the rates of thyroid cancer (AAPC = 3.8; CI = 1.8, 5.8). Incidence continues to remain low relative to other cancers 2,3 and rates are not calculable for the period between 1986 and Comparable increases in thyroid cancer are evident at the provincial and national levels. This increase is most likely a consequence of the improvement in detection technologies such as ultrasound and needle biopsy that have led to identifying early stage tumours more frequently. Overall, there was no statistically significant increase in breast cancer incidence (AAPC = 0.4; CI -0.4, 1.2). From figure 6a, there was a steady increase in breast cancer incidence between 1986 and 1998, but since there has been a steady decrease. Changing reproductive histories (women choosing to bear children later in life), and the introduction of screening programs including mammography may partially account for the increase in rates between 1986 and 1998, but better awareness and change in lifestyle may account for the decrease over the last 5 years. Colorectal cancer is the third most common cancer in women. While the number of new cases rose by 64% (a decrease of 17% from the period ), the age-standardized incidence rates remained stable during the period considered in this report. Cancer of the uterus is the fourth most common cancer in women in the Niagara Region and the most common malignancy of the female genital system. Between 1986 and 2003 there were a total of 1126 cases of uterine cancer. Standardized incidence rates in the Region remained stable. 2 Incidence rates for thyroid cancer were calculated for the period between 1986 and Due to the very low incidence rates thyroid cancer trends are not included in the graphs 6a & 6b. 22

23 Figure 6a. Age-standardized incidence rates (ASIR) for selected cancer sites, females, Niagara Region, Age-standardized incidence rates/100, Year Data Source: SEERStat, 2006 Breast Colorectal Lung Non-Hodgkin's Lymphoma Uterine Rates Standardized to 1991 Canadian Population 23

24 Mortality As a result of improved detection and treatment methods, stable or increasing rates in incidence have not generally been accompanied by an increase in mortality rates. Of all deaths from cancers among women, only Non-Hodgkin s Lymphoma mortality has risen significantly (AAPC = 2.5; CI = 0.1, 5) and the only noteworthy decline was in breast cancer (AAPC = -1.7; CI = -2.6, -0.8). Treatment and screening methods for breast cancer have improved greatly and may be the reason for the reduction in mortality. Figure 6 b. Age-standardized mortality rates for selected cancer sites, females, Niagara Region, Age-standardized mortality rates/100, Year Data Source: SEERStat, 2006 Breast Lung Colorectal Non-Hodgkin's Lymphoma Uterine Rates Standardized to 1991 Canadian Population 24

25 C. Average Annual Percentage Change in Incidence and Mortality of Selected Cancers (Niagara Region, ) Table 2. Average annual percentage change in incidence and mortality (Niagara Region, ) AAPC Incidence AAPC Mortality Male Female Male Female All Cancers * -0.7* -0.1 Oral -2.8* ~ Stomach -2.7* -4.0* -2.8* -4.5* Colorectal * -0.2 Pancreas Lung -1.6* 1.3* -1.9* 1 Melanoma of the Skin ~ Breast n/a 0.4 ~ -1.7* Cervix n/a -0.9 ~ ~ Uterus n/a 0 ~ ~ Ovary n/a 1.7 ~ 1.2 Prostate 1.8* n/a -0.9 n/a Urinary Bladder -2.4* -0.9 ~ ~ Kidney ~ Thyroid ~ ~ ~ ~ Hodgkin Lymphoma ~ ~ ~ ~ Non-Hodgkin Lymphoma 1.8* 2.2* 2.5* 2.5* Leukemia * 0.4 n/a Not applicable Data Source: SEERStat, 2006 * Significant at p < 0.05; note that statistical significance in AAPC refers to an increase or decrease over time not to differences between males and females. ~ low numbers, statistic could not be calculated Due to changes in bladder cancer classification in 1988, bladder cancer is not included in this table. 25

26 D. Comparison of Niagara Region with Provincial Temporal Trends in Incidence and Mortality This section describes the 10 most common cancers in males and females living in the Niagara Region and provides a comparison with provincial temporal trends. Prostate Cancer The incidence of cancer of the prostate increased steadily in the Niagara Region at an average annual rate of approximately 3.8%, from 80 cases per 100,000 (1986) to 119 cases per 100,000 (2003) (Figure 7). In the 5-year period between 1988 and 1993, the rate of prostate cancer increased by 60%. This spike in cases diagnosed is at least partly due to the detection of cancers following trans-urethral resection of the prostate (TURP) for suspected benign prostatic hypertrophy in addition to improved detection as a result of the Canada-wide introduction of prostate-specific antigen test in While an increase in rates is also evident at the provincial level, regional rates appear to be slightly lower than Ontario s between 1986 and Since 1997 rates have not varied significantly. The only fully established risk factors for prostate cancer are increasing age, African- American ethnicity, and family history of the disease. Research has almost conclusively established a link for male hormones (androgens) in the causation of prostate cancer, but the effects of these hormones are not fully understood. Although no lifestyle factors (including diet and exercise) have been conclusively established as prostate cancer risk or protective factors, there is some evidence that a diet high in animal fat may account for differences noted in the incidence between different countries (NCI a, 2007). Localized prostate cancer is normally treated with radical prostatectomy or radiation therapy. There are 2 tests available to detect prostate cancer in the absence of symptoms, the digital rectal exam (DRE) and a blood test for prostate-specific antigen (PSA). However, conditions other than prostate cancer may result in elevated levels of PSA. Not all men with moderately elevated levels of PSA have prostate cancer, and some men with prostate cancer have normal levels of PSA. A combined approach using PSA and DRE is most valuable in detecting prostate cancer. However, a biopsy is the method to definitively diagnose cancer of the prostate (NCI b, 2007). Although PSA is currently the best serum marker for prostate cancer, there is no evidence presently from random clinical trials that screening of prostate cancer using PSA decreases mortality (Armstrong, 2002). RESULTS 26

27 Figure 7. Temporal trends of Prostate Cancer, , Niagara Region and Ontario Age-standardized rate/100, Year Niagara Incidence Ontario Incidence Niagara Mortality Ontario Mortality Data Source: SEERStat, 2006 Rates Standardized to 1991Canadian Population 27

28 Breast Cancer In the Niagara Region (NR), breast cancer results in the largest number of new cancer cases in women, in addition to being the one of the primary causes of cancer mortality. Between 1986 and 2003 significant increasing time trends are evident in the incidence of breast cancer (Figure 8). During this period there was a 13% increase in the incidence rate, but this was not statistically significant (APC= 0.4; CI = -0.4, 1.2). The increase may be due in part to the rising number of screening mamographies, in addition to changing reproductive histories as a result of bearing children at a later age. While the incidence of breast cancer rose, the rate of women dying from the disease declined by approximately 19% from 33.6/100,000 to 27.2/100,000 (APC = -1.7; CI = -2.6, -0.8). The likelihood of a woman developing breast cancer increases rapidly with age, making age the most significant risk factor for breast cancer. Other risk factors include family history of breast cancer, a history of breast cancer in one breast, a history of certain types of benign breast disease, and high levels of radiation exposure to the chest. Other wellestablished, but weak risk factors include obesity in post-menopausal women and various reproductive risk factors (age at first parity 30+ years of age), early onset of menstruation, late onset of menopause). Demographic factors that increase the risk of developing breast cancer include living in an urban area, belonging to a higher socioeconomic class, and being born in North America or Northern Europe (NCI c, 2007). Behavioural risks: the use of post-menopausal estrogen replacement therapy, high intake of dietary fat, alcohol use and physical inactivity have been linked to the risk of breast cancer (NCI c, 2007). Early detection through mammography in women over the age of 50 has been shown to reduce mortality by approximately 30%. A breast-cancer screening programme is established in Ontario for women years. 28

29 Figure 8. Temporal trends for Breast Cancer, , Niagara Region and Ontario Age-standardized rate/100, Year Data Source: SEERStat, 2006 Niagara Incidence Ontario Incidence Niagara Mortality Ontario Mortality Rates Standardized to 1991Canadian Population 29

30 Lung Cancer Among women in the Niagara Region, lung cancer incidence rates have increased significantly, approximately 1.3% per year since 1986 (CI: 0, 2.6) (Figure 9). With increasing incidence, the mortality rates have remained stable. When comparing women to men, the rates in women are nearly half of those in men but incidence rates in men are decreasing at approximately 1.7 % per year (CI: -2.5, -0.9) (Figure 10). Mortality in men has also been decreasing at approximately 1.9% per year (CI: -3.1, -0.7), while the mortality rates in women have remained stable. The patterns for both men and women in the Niagara region follow very similar patterns to Ontario for both incidence and mortality. Lung cancer includes tumours of the trachea, bronchi and lung. Virtually all lung cancers arise in epithelial tissue, and most originate from the lining of the bronchi. Active and passive exposure to tobacco smoke is the foremost cause of lung cancer, but a small proportion is the result of occupational exposures to known asbestos and benzene. Longterm exposure to particulate matter (air pollution) increases the mortality risk of lung cancer (Jerrett et al., 2005; Krewski et al, 2005; Pope et al., 2002). There is a strong incentive to identify an appropriate screening test for lung cancer, since once it s diagnosed the prognosis is poor [5 year relative survival ratio: 15% (Canadian Cancer Statistics, 2007)]. However, there are no prescribed screening programmes for lung cancer since evidence indicates that screening does not impact the risk of fatal lung cancer. Also, lung cancer screening leads to a high number of false positives, thus an increase in invasive follow-up testing (Canadian Cancer Society, 2006; Armstrong, 2002; Fontana et al., 1996). Reduction/elimination of cigarette smoking would prevent the majority of lung cancers. 30

31 Figure 9. Temporal trends of Lung Cancer in females, , Niagara Region and Ontario Age-standardized rate/100, Year Data Source: SEERStat, 2006 Niagara Incidence Ontario Incidence Niagara Mortality Ontario Mortality Rates Standardized to Canadian 1991 Population Figure 10. Temporal trends in Lung Cancer in males, , Niagara Region and Ontario Age-standardized Rate/100, Year Source: SEERStat, 2006 Niagara Incidence Ontario Incidence Niagara Mortality Ontario Mortality Rates Standardized to 1991Canadian Population Data 31

32 Colorectal Cancer While there hasn t been a statistically significant change in the incidence of colorectal cancer in women (Figure 11) or men (Figure 12) in the Niagara Region, the downward trendline parallels that of Ontario s decreasing trends. In the Niagara Region, the mortality rate for men has significantly decreased at approximately 1.6% per year between 1986 and 2003 (CI = -2.6,-0.6), while the mortality rate for females has remained relatively stable.(figures 11 & 12). Regional incidence rates for colorectal cancer in women residing in Niagara is slightly lower than provincial rates. Risk factors for colorectal cancer include: 1. Age, most people who have colorectal cancer are over age 50; 2. A high fat, high calorie and low fibre diet; 3. lack of physical activity; 4. Certain kind of polyps are believed to lead to colorectal cancer; 5. Personal history-people who have had colorectal cancer, as well as ovarian, uterine, or breast cancers, have a slightly increased risk for colorectal cancer; 6. Family history-individuals with first-degree relatives who have had colorectal cancer have an increased risk for colorectal cancer; and 7. Ulcerative colitis-people with ulcerative colitis (inflamed lining of the colon) (WCRF-AICR, 1997; WCRF-AICR, 2007; NCI d, 2007). Stool markers and virtual colonoscopy are potential screening modes (Armstrong, 2002). In the United States randomized controlled trials (RCT) have shown the efficacy of screening for colorectal cancer using the fecal occult blood test and follow-up with colonoscopy (NCI e, 2007). In Spring 2007 a population based colorectal screening program will be implemented in Ontario (CCO, 2007 a ). With this program, we may start seeing a decrease in mortality, as cancers may be caught earlier, thus being able to start treatment earlier. (CCO, 2007 b ). 32

33 Figure 11. Temporal trends in Colorectal Cancer in females, , Niagara Region and Ontario Age-standardized rate/100, Year Data Source: SEERStat, 2006 Niagara Incidence Ontario Incidence Niagara Mortality Ontario Mortality Rates Standardized to 1991 Canadian Population Figure 12. Temporal trends in Colorectal Cancer in males, , Niagara Region and Ontario Age-standardized rate/100, Year Data Source: SEERStat, 2006 Niagara Incidence Ontario Incidence Niagara Mortality Ontario Mortality Rates Standardized to 1991Canadian Population 33

34 Stomach Cancer RESULTS Niagara Region incidence rates of stomach cancer are declining in both women (AAPC = -4.0; CI = -6.4, -1.6) and men (AAPC = -2.6; CI = -4.1, -1.2)(Figure 13 & 14). Niagara Region is closely following the decreasing pattern of incidence in Ontario. Mortality rates of stomach cancer in both men (AAPC = -2.8; CI = -4.9, -0.7) and women (AAPC = -4.5; CI: -7.2, -1.7) continue to show a prominent decline, also reflecting trends in provincial rates. The most common form of cancer affecting the stomach is adenocarcinoma, (arising from the innermost layer of the stomach). The tumour tends to spread through the wall of the stomach and from there into the adjoining organs (pancreas and spleen) and lymph nodes. Factors that increase an individual s risk of contracting the disease are: 1. Long term infection with Heliobacter pylori; 2. A diet high in smoked foods, salted fish and meat, and pickled vegetables. On the other hand, eating whole grain products, fresh fruits and vegetables that contain vitamin A and C appear to lower the risk; 3. Tobacco use; 4. previous stomach surgery; 5. Pernicious anemia; 6. Menetrier s disease (also known as hypertrophic gastropathy); 6. Gender-men are more likely to be affected by stomach cancer than women; 7. ethnicity-the rate is higher in Hispanics and African Americans than in non-hispanic whites. The highest rates are seen in the Asian/Pacific Islanders; 8. age-there is a sharp increase in stomach cancer after the age of 50 years; 9. Type A Blood-the reason for this is unknown; 10. familial cancer syndromes and a family history of stomach cancer (Mayo Clinic a, 2004). There is evidence from Japan, where the incidence of gastric cancer is high, that screening of individuals over 40 years of age with barium x-ray increases survival rates. In Canada, the incidence of stomach cancer is relatively low and there are presently no screening guidelines (Armstrong, 2002). Investigations are underway into potential screening tools: genetic mutations to identify high risk patients, tumour markers, and the eradication of H. pylori (Armstrong, 2002). 34

35 Figure13. Temporal trends in Stomach Cancer in females , Niagara Region and Ontario Age-standardized rate/100, Year Data Source: SEERStat, 2006 Niagara Incidence Ontario Incidence Niagara Mortality Ontario Mortality Rates Standardized to 1991Canadian Population Figure 14. Temporal trends in Stomach Cancer in males, , Niagara Region and Ontario Age-standardized rate/100, Year Data Source: SEERStat, 2006 Niagara Incidence Ontario Incidence Niagara Mortality Ontario Mortality Rates Standardized to 1991Canadian Population 35